I strongly think the PFS endpoint blew up somehow years ago and can not be fixed.
On the OS endpoint I have been unable to form a real opinion. I would not be surprised to see it go either way. I think it will "formally" not be stat sig once PFS fails, but I acknowledge that the FDA can go with a clear OS result despite a technically failed trial.
As I said in a post to Senti a couple years ago. If this was a pure play on the "science" w/o having the extra risk of what the hell LP is really telling you, I might take a flyer on it.